Software Engineer, Ori

Role overview Join a focused team building client-side AI tooling that lets any large language model plug into terminals, chat workspaces, and desktop apps. It is a hands-on individual contributor position with broad surface area, where the engineer will help grow a product line aimed at putting models to work for enterprise users beyond just engineering teams.

Responsibilities - Build and ship across multiple product surfaces, including a CLI and its integrations with third-party coding agent harnesses. - Make models interchangeable across harnesses by enabling model switching, provider fallbacks, per-model tool and prompt tuning, and respecting organizational guardrails without extra user authentication. - Develop the underlying agent runtime, covering sandboxed execution, secrets handling, session management, and shared, inheritable agent skills. - Move agents from local machines to the cloud, supporting on-demand deployment, forking, and shutdown with auditable, version-controlled state. - Own end-to-end quality, from install flow and CLI ergonomics through reliability, observability, and evaluation pipelines. - Surface gaps back into shared APIs and SDKs so that the broader ecosystem can build multi-model agents on common primitives.

Requirements - 5+ years of experience shipping production software, with real ownership of developer tools, CLIs, agent products, or the infrastructure behind them. - Daily fluency with coding agents such as Claude Code, Codex, OpenCode, Pi, or Hermes, plus strong opinions on what makes agent loops and tool integrations work well. - Solid grasp of how agents function internally, including tool calling, context management, prompt caching, evals, and the cost, latency, and quality tradeoffs between models. - Strong systems judgment for safely running agents at scale, including sandboxing, process isolation, secrets management, and reliability across many users. - High agency and a bias toward action, with comfort prototyping, shipping quickly, and iterating on a small team with fluid boundaries. - Product taste for developer tools, with attention to interaction details and how the first five minutes feel for new users.

Nice to have - Built or maintained an open-source agent harness, CLI, or developer tool. - Experience with multi-agent systems, agent skills or memory, and evaluation or benchmarking frameworks. - Shipped a Slack app or a native macOS application. - Existing user of comparable AI tooling, or active side projects in the AI tooling space.